maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[mlir][affine] Rigorous check for loop unrolling store operation 

Static loop unrolling does not change the operation type. We can rigorously make sure to use affine.store in the check.

Reviewed By: ftynse

Differential Revision: https://reviews.llvm.org/D128237
This commit is contained in:
lewuathe 2022-06-25 08:59:13 +09:00
parent 96f6ec5090
commit bd861056a5
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
mlir/test/Dialect/SCF/loop-unroll.mlir
View File

@ -271,7 +271,7 @@ func.func @static_loop_unroll_up_to_factor(%arg0 : memref<?xf32>) {
// UNROLL-UP-TO-DAG: %[[C0:.*]] = arith.constant 0 : index
// UNROLL-UP-TO-DAG: %[[C2:.*]] = arith.constant 2 : index
// UNROLL-UP-TO-NEXT: %[[V0:.*]] = affine.apply {{.*}}
// UNROLL-UP-TO-NEXT: store %{{.*}}, %[[MEM]][%[[V0]]] : memref<?xf32>
// UNROLL-UP-TO-NEXT: affine.store %{{.*}}, %[[MEM]][%[[V0]]] : memref<?xf32>
// UNROLL-UP-TO-NEXT: %[[V1:.*]] = affine.apply {{.*}}
// UNROLL-UP-TO-NEXT: affine.store %{{.*}}, %[[MEM]][%[[V1]]] : memref<?xf32>
// UNROLL-UP-TO-NEXT: return